The present invention relates in general to mechanisms for coupling two components, and in specific to connectors for coupling microcomponents.
Extraordinary advances are being made in micromechanical devices and microelectronic devices. Further, advances are being made in MicroElectroMechanical (xe2x80x9cMEMsxe2x80x9d) devices, which comprise integrated micromechanical and microelectronic devices. The terms xe2x80x9cmicrocomponentxe2x80x9d and xe2x80x9cmicrodevicexe2x80x9d will be used herein generically to encompass microelectronic components, micromechanical components, as well as MEMs components. A need exists in the prior art for a mechanism for coupling microcomponents. For example, a need exists for some type of mechanical connector that provides either a permanent mechanical coupling or a temporary coupling between two or more microcomponents.
Generally, microcomponent devices are devices having a size below one millimeter by one millimeter. Although, microcomponents as large as one centimeter by one centimeter have been provided in the prior art. Moreover, microcomponents may be smaller than one millimeter by one millimeter in size. Furthermore, techniques for fabricating microcomponents typically produce such microcomponents having a minimum feature size of approximately one micron. Although, such microcomponents may be implemented with a minimum feature size of greater or less than one micron.
Various types of coupling mechanisms are well known for large scale assembly. For example, mechanisms such as screws, bolts, rivets, snap connectors, and a variety of other types of coupling mechanisms are well known and commonly used for coupling large scale components. However, such coupling mechanisms for large scale components are very difficult to implement on the small scale necessary for coupling microcomponents. That is, many large scale coupling mechanisms are unacceptable and are not easily adaptable for coupling microcomponents.
Microcomponents, such as MEMs, are generally fabricated as two dimensional (xe2x80x9c2-Dxe2x80x9d) components. That is, microcomponents generally have a defined 2-D shape (e.g., defined X dimension and Y dimension), but the third dimension (e.g., the Z dimension) is generally set for the entire part. Limited control over the Z dimension may be achieved by using multiple layers in microcomponent designs. Although, any given layer of the microcomponent is a given thickness. Thus, a more desirable method to alter the Z dimension, is to combine microcomponent parts together.
One prior art technique commonly used for assembling microcomponents, such as MEMs, is serial microassembly, which may also be referred to as xe2x80x9cpick and placexe2x80x9d assembly. With serial microassembly, each device is assembled together one component at a time, in a serial fashion. For example, if a device is formed by combining two microcomponents together, a placing mechanism is used to pick up one of the two microcomponents and place it on a desired location of the other microcomponent. While such a serial microassembly technique using pick and place operations initially appears to be a simple technique, when working with microcomponents, such pick and place operations are very complex. For microassembly, the relative importance of the forces that operate is very different from that in the macro world. For example, gravity is usually negligible, while surface adhesion and electrostatic forces dominate. (See e.g., A survey of sticking effects for micro parts handling, by R. S. Fearing, IEEE/RSJ Int. Workshop on Intelligent Robots and Systems, 1995; Hexsil tweezers for teleoperated microassembly, by C. G. Keller and R. T. Howe, IEEE Micro Electro Mechanical Systems Workshop, 1997, pp. 72-77; and Microassembly Technologies for MEMS, by Micheal B. Cohn, Karl F. Bxc3x6hringer, J. Mark Noworolski, Angad Singh, Chris G. Keller, Ken Y. Goldberg, and Roger T. Howe). Due to scaling effects, forces that are insignificant at the macro scale become dominant at the micro scale (and vice versa). For example, when parts to be handled are less than one millimeter in size, adhesive forces between a gripper and an object can be significant compared to gravitational forces. These adhesive forces arise primarily from surface tension, van der Waals, and electrostatic attractions and can be a fundamental limitation to handling of microcomponents. While it is possible to fabricate miniature versions of conventional robot grippers in the prior art, overcoming adhesion effects for such small-scale components has been a recognized problem.
Often in attempting to place a microcomponent in a desired location, the component will xe2x80x9cstickxe2x80x9d or adhere to the placing mechanism due to the aforementioned surface adhesion forces present in microassembly, making it very difficult to place the component in a desired location. (See e.g. Microfabricated High Aspect Ratio Silicon Flexures, Chris Keller, 1998). For example, small-scale xe2x80x9ctweezersxe2x80x9d (or other types of xe2x80x9cgrippersxe2x80x9d) are used to perform such pick and place operations of serial microassembly, and often a component will adhere to the tweezers rather than the desired location, making placing the component very difficult. It has been recognized in the prior art that to grip microcomponents and then attach them to the workpiece in the desired orientation, it is essential that a hierarchy of adhesive forces be established. For instance, electrostatic forces due to surface charges or ions in the ambient must be minimized. Adhesion of the micropart to the unclamped gripper surfaces (with zero applied force) should be less than the adhesion of the micropart to the substrate, to allow precise positioning of the part in the gripper.
Accordingly, unconventional approaches have been proposed for performing the pick and place operations. For example, Arai and Fukada have built manipulators with heated micro holes. See A new pick up and release method by heating for micromanipulation, by F. Arai and T. Fukada, IEEE Micro Electro Mechanical Systems Workshop, 1997, pp. 383-388). When the holes cool, they act as suction cups whose lower pressure holds appropriately shaped objects in place. Heating of the cavities increases the pressure and causes the objects to detach from the manipulator. Alternatively, some type of external adhesive (e.g., a type of liquid xe2x80x9cgluexe2x80x9d) may be utilized to enable the microcomponent to be placed in a desired location. That is, because the components themselves provide no mechanism for coupling, an external adhesive may be required to overcome the adhesive force between the component and the placing mechanism (e.g., tweezers). For example, the target spot on the workpiece may have a surface coating that provides sufficiently strong adhesion to exceed that between the micropart and the unclamped gripper.
Another prior art technique commonly used for assembling microcomponents, such as MEMs, is parallel microassembly. In parallel microassembly, components of one wafer are coupled to components of another wafer simultaneously in a single step. For example, the above pick and place operations may be performed on an entire wafer, such that one wafer is picked up and placed onto another wafer, thereby coupling the components of one wafer with the components of the other wafer. Therefore, parallel assembly involves the simultaneous precise organization of an ensemble of microcomponents. This can be achieved by microstructure transfer between aligned wafers or arrays of binding sites that trap an initially random collection of parts. Binding sites can be micromachined cavities or electrostatic traps; short-range attractive forces and random agitation of the parts serve to fill the sites.
Parallel microassembly techniques may be categorized as either xe2x80x9cdeterministicxe2x80x9d or xe2x80x9cstochastic,xe2x80x9d depending on whether the microcomponents are initially organized. There are two general approaches to parallel microassembly in the prior art, one based on the massively parallel transfer between wafers of arrays of microcomponents (i.e., xe2x80x9cdeterministic parallel microassemblyxe2x80x9d) and one utilizing various approaches to orient an initially random array of microcomponents (i.e., xe2x80x9cstochastic parallel microassemblyxe2x80x9d). Deterministic parallel microassembly refers to direct, wafer-to-wafer transfer of microcomponents. Since the placement of the microcomponents is predetermined by their layout on the donor wafer, the challenge with such process typically lies in bonding the components to the target. A common technique for bonding the components utilizes solder bumps to achieve such bonding.
While conventional assembly techniques have been successfully adapted from the macro world, the molecular regime offers many examples of efficient assembly processes. Crystal growth, antibody-antigen recognition, and most other chemical and biological behaviors are mediated by thermal motion and interparticle forces. In contrast to the macroscopic concepts of manipulators and path planning, a molecular system may be analyzed as an ensemble of particles evolving toward a state of minimal potential energy. The lure of this thermodynamic approach is that when parts must be redistributed or reoriented, a single complex manipulator may be replaced by an array of lithographically-defined binding sites. Such sites might consist of electrostatic traps, or simply etched wells on a substrate. Thermodynamic analysis shows the potential for massively parallel operation forming assemblies 106 or more elements in seconds, with placement tolerance limited by lithographic accuracy. Historically, stochastic assembly precedes MEMS by several decades. One of the best illustrations is the work of Yando disclosed in U.S. Pat. No. 3,439,416 entitled xe2x80x9cMethod and Apparatus for Fabricating an Array of Discrete Elementsxe2x80x9d issued in 1969, which discloses an array of magnets on which particles with magnetic coatings are placed, vibrated, and trapped so as to form a matching array. Each particle is described as a microelectronic device, such as a diode. One problem with this scheme is that the magnet arrays are composed of laminated sheets stacked perpendicularly to the place of the array, so that many laminations are needed to achieve an array of appreciable extent. A further example is the APOS parts feeder described by Hitakawa. (See Advanced Parts Orientation System Has Wide Application, by H. Hitakawa, Assembly Automation, 8(3), 1988). The feeder uses an array of xe2x80x9cberthsxe2x80x9d cut into a vibrating plate. Parts are fed over the plate, and the berths are designed, like the track of the bowl feeder, to accept only parts in a given orientation. Eventually, all the berths are filled. Various stochastic assembly xe2x80x9cwet processesxe2x80x9d have also been proposed. (See e.g., Fluidic Self-assembly of Microstructures and its Application to the Integration of GaAs on Si, by H. J. Yeh and J. S. Smith, Proceedings IEEE Micro Electro Mechanical Systems, Oiso, Japan, Jan. 25-28, 1994/New York: IEEE, 1994, p. 279-84; and Self-Orienting Fluidic Transport (SOFT) Assembly of Liquid Crystal Displays, by M. A. Hadley, presentation at the Defense Manufacturing Conference, Palm Springs, Calif., Dec. 1-4, 1997). Additionally, various stochastic assembly xe2x80x9cdry processesxe2x80x9d have been proposed. For example, in 1991, Cohn, Kim, and Pisano described stochastic assembly using vibration and gravitational forces to assemble arrays of up to 1000 silicon chiplets. (See Self-Assembling Electrical Networks: An Application of Micromachining Technology, by M. Cohn, C. J. Kim, and A. P. Pisano, Transducers 91 International Conference on Solid-State Sensors and Actuators, Digest of Technical Papers, San Francisco, Jun. 24-27, 1991/New York: IEEE, 1991, p. 493).
Snap connectors have been proposed in the prior art as a mechanism for coupling MEMs components (see e.g., Design, Fabrication, and Characterization of Single Crystal Silicon Latching Snap Fasteners for Micro Assembly, by Rama Prasad, Karl-Friedrich Bxc3x6hringer, and Noel C. MacDonald, Proc. ASME Int. Mech. Eng. Congress and Exposition, 1995). A snap connector as proposed in the prior art is shown in FIGS. 9A and 9B. As shown in FIG. 9A, a snap connector 940 having an xe2x80x9canchorxe2x80x9d (or xe2x80x9cbarbed endxe2x80x9d) 942 is coupled to a component 970. Furthermore, a mating component 960 coupled to a different component 980 is provided, which includes xe2x80x9clatchesxe2x80x9d 962 and 964. The snap connector 940 of the prior art is moved linearly along the plane of the wafer (i.e., along the X axis of FIGS. 9A and 9B) to enable the component 970 to be coupled to the component 980 within the plane of the wafer. FIG. 9B illustrates such a coupling.
Thus, snap connectors have been proposed that are capable of coupling MEMs components. However, the snap connectors of the prior art are designed to only work in the plane of the wafer on which the components are located (i.e., along the X and Y axes of FIGS. 9A and 9B. Thus, the snap connectors of the prior art do not provide a mechanism for general assembly, but rather are only capable of coupling components in the plane of the wafer. Thus, for example, no mechanism has been disclosed in the prior art for using a snap connector for connecting a component perpendicular to the surface of a wafer to achieve 3-D assembly. That is, prior art snap connector implementations do not provide a mechanism suitable for general coupling of two microcomponents, but rather only allow for coupling of microcomponents in-plane (i.e., to achieve only 2-D assembly).
Additionally, the in-plane snap connector of the prior art lacks many characteristics that may be desired for a general coupling mechanism. For example, as shown in FIG. 9B, the snap connector works to prevent the components 970 and 980 from becoming uncoupled by a movement within the plane of the wafer (i.e., by a movement in the X or Y direction of FIGS. 9A and 9B). Although, the mating component 960 includes area 950, which permits a certain amount of xe2x80x9cplayxe2x80x9d between the components along the X axis of FIGS. 9A and 9B. Thus, the prior art snap connector does not constrain undesirable motion/movement between the coupled components. That is, nothing prevents snap connector 940 from proceeding further inward along the X axis within the mating component 960, and thereafter the snap connector 940 may proceed outward along the X axis within the mating component 960 until the latches 962 and 964 engage the barbed end 942. Furthermore, in such prior art snap connector implementation nothing prevents the snap connector 940 from moving out of the plane of the wafer (i.e., along the Z axis of FIGS. 9A and 9B), thereby permitting the components 970 and 980 to become uncoupled. Thus, the prior art snap connector only enables 2-D assembly in which components are coupled only in two dimensions, and does not provide a connector suitable for general assembly, which may include use for performing 3-D assembly. The prior art connector requires the translator positioning component 970 relative to component 980 to exert the force required to extend cantilever springs 960 and 966. These, as well as other characteristics of the prior art 2-D snap connectors make such prior art snap connectors unsuitable for general (or xe2x80x9call-purposexe2x80x9d) assembly using low or near-zero insertion force translators.
Also, xe2x80x9csnap locksxe2x80x9d have been proposed for use in assembling a hollow triangular beam (see e.g., Surfaced-Micromachined Components for Articulated MicroRobots, by Richard Yeh, Ezekiel J. J. Kruglick, and Kristofer S. J. Pister, Journal of MicroElectroMechanical Systems, Vol. 5, No. 1, March 1996). Such a prior art snap lock proposal is shown in FIGS. 10A and 10B. As shown in FIG. 10A, a microcomponent is provided, which includes plates 10, 12, and 14. Plates 10 and 12 are rotatably coupled with a scissor hinge 16, and plates 12 and 14 are also coupled with a scissor hinge 18. Plate 10 includes snap locks 20, which may couple into mating apertures 22 of plate 14. Thus, the 13 hollow triangular beam of FIG. 10B may be formed by rotating plate 10 and 14 upward and coupling plates 10 and 14 with snap locks 20 and mating apertures 22. Accordingly, a 3-D object is formed by assembling the three-plated microcomponent together.
However, the prior art does not teach that such snap locks are suitable for general assembly. Rather, the snap locks are used together to enable a MEMs component to assemble itself to form a hollow triangular beam, but the prior art does not teach how such snap locks may be utilized for general assembly in coupling two separate components together. Additionally, it appears that such snap lock of the prior art may lack many characteristics that may be desired for a general coupling mechanism. For example, such snap lock does not allow for coupling a component normal to the wafer surface, but instead three hinged plates are disclosed such that two plates may be coupled at approximately 60 degrees to the wafer surface. Thus, the coupled plates 10 and 14 assist in maintaining the coupling by xe2x80x9cleaningxe2x80x9d on each other. No teaching suggests that the snap lock disclosed may be used for a general assembly operation that does not permit such xe2x80x9cleaning,xe2x80x9d such as two microcomponents being coupled normal to each other. The pointed end 20 of the snap locks must deform during assembly. The force required for deformation needs to be provided to all xe2x80x98arrowheadsxe2x80x99 simultaneously, making this a high-insertion force connector. As a result, the prior art teaching does not disclose a snap connector suitable for general assembly of microcomponents using near-zero insertion force, but rather provides only specific purpose mechanisms for assembling one particular type of MEMs component.
In view of the above, a desire exists for a coupling mechanism suitable for the assembly of microcomponents. A particular desire exists for a coupling mechanism that is suitable for performing general assembly of microcomponents, including 3-D assembly. A further desire exists for a coupling mechanism that enables microcomponents to be securely coupled to each other in a manner that constrains undesirable movement of the coupled components relative to each other.
These and other objects, features and technical advantages are achieved by a system and method which provide a general-purpose connector suitable for coupling microcomponents. More specifically, a snap connector is disclosed that is suitable for performing general assembly, including out-of-plane, 3-D assembly of microcomponents, wherein such microcomponents may be securely coupled together. That is, connectors are disclosed which enable microcomponents to be coupled in a manner that constrains undesirable movement of the coupled components relative to each other. Preferably, the connectors enable assembly operations utilizing a near-zero insertion force translator. For example, in one embodiment a sufficient force is provided by a relatively high force gripper to compress or xe2x80x9cpreloadxe2x80x9d engaging members (e.g., xe2x80x9carmsxe2x80x9d) of a connector to a desired position for engaging another component, wherein such engagement can be performed by a relatively small or near-zero force translator.
A preferred embodiment provides a xe2x80x9cpreloadedxe2x80x9d snap connector that may be utilized to perform general assembly of microcomponents. An alternative embodiment provides a non-preloaded snap connector suitable for performing general assembly of microcomponents. Still a further alternative embodiment provides a xe2x80x9csqueezexe2x80x9d snap connector that is suitable for performing general assembly of microcomponents. Such connectors may be implemented as an integrated part of a microcomponent, or they may be implemented as separate, stand-alone correctors. For example, a dual-ended snap connector is disclosed herein, which may be coupled to a first microcomponent, and then used to couple the first microcomponent to a second microcomponent.
The foregoing has outlined rather broadly the features and technical advantages of the present invention in order that the detailed description of the invention that follows may be better understood. Additional features and advantages of the invention will be described hereinafter which form the subject of the claims of the invention. It should be appreciated by those skilled in the art that the conception and specific embodiment disclosed may be readily utilized as a basis for modifying or designing other structures for carrying out the same purposes of the present invention. It should also be realized by those skilled in the art that such equivalent constructions do not depart from the spirit and scope of the invention as set forth in the appended claims.